Background. Artemisinin-based combination therapy (ACT) reduces microscopically confirmed gametocytemia and mosquito infection. However, molecular techniques have recently revealed high prevalences of submicroscopic gametocytemia. Our objective here was to determine the effect of sulfadoxine-pyrimethamine (SP) monotherapy and treatment with SP plus amodiaquine (AQ), SP plus artesunate (AS), and artemether-lumefantrine (AL; Coartem) on submicroscopic gametocytemia and infectiousness.Methods. Kenyan children (n=528) 6 months-10 years of age were randomized to 4 treatment arms. Gametocytemia was determined by both microscopy and Pfs25 RNA-based quantitative nucleic acid sequence-based amplification (Pfs25 QT-NASBA). Transmission was determined by membrane-feeding assays.Results. Gametocyte prevalence, as determined by Pfs25 QT-NASBA, was 89.4% (219/245) at enrollment and decreased after treatment with SP plus AS, SP plus AQ, and AL. Membrane-feeding assays for a group of randomly selected children revealed that the proportion of infectious children was as much as 4-fold higher than expected when based on microscopy. ACT did not significantly reduce the proportion of infectious children but did reduce the proportion of infected mosquitoes.Conclusions. Submicroscopic gametocytemia is common after treatment and contributes considerably to mosquito infection. Our findings should be interpreted in the context of transmission intensity, but the effect of ACT on malaria transmission appears to be moderate and restricted to the duration of gametocyte carriage and the proportion of mosquitoes that are infected by carriers.